Machine Learning Engineer at JPMorgan Chase & Co.

Built and optimized machine learning models for financial analytics, improving prediction accuracy by 35% and enabling data-driven decisions. Developed LSTM-based fraud detection models with 98% accuracy, significantly enhancing compliance. Improved text classification by fine-tuning BERT and using GAN-generated data. Established CNNs for anomaly detection to reduce fraudulent activities by 38%. Conducted hyperparameter tuning, feature engineering, and built robust inference pipelines with FastAPI deployed on AWS services. Leveraged cloud ML platforms to speed development and deployment by 30%, reducing latency by 40%. Collaborated with analytics and engineering teams to design scalable ML pipelines and predicted peak hours with time-series models, resulting in 30% cost reduction during off-peak periods. Built high-performance backends using FastAPI and multiprocessing for real-time monitoring and 2x speed improvement.

AI & Deep Learning Developer at Aspire Technolab

Designed and fine-tuned ML models like BERT and RNN-based NER for intent classification and key entity extraction with high accuracy. Developed automated pipelines leveraging OCR, RAG, and OpenCV to increase text extraction accuracy by 98%. Optimized algorithms to boost model efficiency by 38%, improving anomaly detection and image recognition accuracy. Built and deployed XGBoost, CNN, and time-series models to optimize operations and increase sales. Managed scalable AWS model deployments with reduced issue resolution time. Used libraries like Pandas, NumPy, SciPy, PyTorch, and OpenCV for preprocessing and feature extraction. Applied unsupervised methods such as k-means, GMM, and Encoder-Decoder architectures to enhance text embedding precision by 20%. Collaborated on customer service chatbot development using LLMs integrated with Confluence, improving service satisfaction by 25%. Conducted EDA and developed LSTM models for multi-class sentiment prediction with 89% validation F1-score.

Machine Learning Engineer at JPMorgan Chase & Co.

Built and optimized ML models for financial analytics, achieving a 35% improvement in prediction accuracy and enabling data-driven decision-making. Engineered LSTM-based fraud detection with a 98% accuracy rate, identifying high-risk transactions and enhancing compliance measures. Fine-tuned BERT for text classification and incorporated data generated by GANs to improve model performance. Implemented CNN-based anomaly detection for financial transactions, reducing fraudulent activities by 38%. Developed robust inference pipelines using FastAPI and deployed models on AWS services (EC2, Lambda, SageMaker) to support high-volume processing. Leveraged cloud platforms like AWS SageMaker and Google Cloud AI Platform to accelerate development and deployment, reducing time-to-market by 30%. Automated ML workflows and improved operational latency by 40%. Partnered with analytics and engineering teams to design scalable ML pipelines for financial products. Used time-series models to predict peak